Non-insulin dependent diabetes mellitus (NIDDM) is a major health problem in the U.S. Diabetes has been diagnosed in approximately seven million Americans, and a similar number may have the disease unknowingly. Over 85% to 90% of all known cases of diabetes in the U.S. are Type II, i.e., non- insulin dependent diabetes mellitus (NIDDM). The importance of NIDDM as a risk factor for periodontitis in older Americans is not known. We hypothesize that non-insulin dependent diabetes mellitus (NIDDM) is an important risk factor for periodontal attachment loss in older persons. Diabetes is a major comorbidity of aging, affecting well over 10% of persons over 65 years old. We propose to carry out a cross-sectional and longitudinal clinical epidemiologic study on the relationship of NIDDM to periodontitis in elders. A better understanding of the role played by NIDDM as a risk factor is essential for the development of strategies to prevent periodontitis and tooth loss in older Americans. Aim l: Periodontal attachment loss will be measured cross-sectionally in elders with NIDDM (N=160) and in medically healthy, non-diabetic elders (N=16O). Using cross-sectional epidemiologic data, we will test the following specific hypotheses: - Prevalence of periodontal disease, as manifested by clinical attachment loss, is greater in elders with NIDDM. - Severity and extent of periodontal disease is greater in elders with NIDDM. - The length of time that elders have been affected by NIDDM is associated with the extent and severity of periodontal disease in elders. Aim 2: We will obtain subgingival plaque samples from mesial sites of teeth in diabetic and non-diabetic subjects with periodontitis. Using DNA probe assays, we will characterize the periodontal microflora in order to ascertain potential microbiologic differences in the periodontitis between diabetic and non-diabetic elders. We will test the null hypothesis that: - The periodontal microflora in diabetic elders with periodontitis is not qualitatively or quantitatively different from that in non-diabetic elders with periodontitis. Aim 3: Periodontal attachment loss changes, over a 12-month interval, will be measured in the elders with NIDDM and in medically healthy elders; and we will test the following hypotheses: - Elders with NIDDM are at higher risk for progression of periodontal disease, and proportionately more will demonstrate evidence of attachment loss over time. - Elders with NIDDM will have more extensive and severe progression of periodontal disease than non-diabetic elders over time.